Consult an unmerged panel of models

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=true and destructiveHint=false, so the safety profile is covered. The description adds meaningful behavioral context: the server never merges/ranks/picks a winner, every model receives the brief byte-for-byte identically, and the output is verbatim and attributed. It also warns that naming a preferred answer primes the panel and destroys the signal. This goes beyond the annotations and helps the agent understand the tool's non-obvious behavior.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is compact and front-loaded: the first sentence states the core behavior, the second gives usage guidance, the third explains the value, and the final sentence clarifies what the tool does NOT do. Every sentence earns its place, and the structure is easy to scan.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a tool with 6 parameters, no output schema, and no nested objects, the description covers the essential behavioral contract: what it sends, what it returns, when to use it, and what it never does. The only minor gap is that it doesn't describe the exact response format (e.g., JSON structure), but since there is no output schema, an agent might benefit from knowing the response shape. However, the description's explicit 'verbatim, attributed, unranked and unsynthesized' gives enough for most use cases.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 50%, so the schema documents brief, models, and system but not max_tokens, timeout_ms, or temperature. The description compensates by explaining the brief's purpose in detail ('Include the goal, hard constraints, current situation, decisions already taken, what was rejected and why, and what is still unknown') and warning against naming a preferred answer. The remaining parameters (max_tokens, timeout_ms, temperature) are standard LLM parameters with clear names and schema constraints, so the gap is minor.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific verb ('Send'), a specific resource ('ONE identical brief to several LLMs in parallel'), and the exact output behavior ('every answer back verbatim, attributed, unranked and unsynthesized'). It also explicitly distinguishes itself from any merging/ranking tool by saying 'This server never merges, ranks or picks a winner.' This is a clear, specific definition that an agent can act on.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description gives explicit when-to-use guidance: 'when a decision is consequential or hard to reverse, when you want your own reasoning stress-tested by models that did not see it, or when the user asks for other opinions or approaches.' It also explains the value of the spread and what a split means. It doesn't name sibling alternatives, but the siblings (get_status, list_models, get_debug_logs) are clearly different operations, so no exclusion is needed.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.